fpsyg-07-00581 April 22, 2016 Time: 16:29 # 1

ORIGINAL RESEARCHpublished: 25 April 2016

doi: 10.3389/fpsyg.2016.00581

Edited by:Radha R. Sharma,

Management Development Institute,India

Reviewed by:M. Teresa Anguera,

University of Barcelona, SpainShane Connelly,

University of Oklahoma, USA

*Correspondence:Cornelia Gerdenitsch

cornelia.gerdenitsch@univie.ac.at

Specialty section:This article was submitted to

Organizational Psychology,a section of the journalFrontiers in Psychology

Received: 13 January 2016Accepted: 08 April 2016Published: 25 April 2016

Citation:Gerdenitsch C, Scheel TE, Andorfer J

and Korunka C (2016) CoworkingSpaces: A Source of Social Support

for Independent Professionals.Front. Psychol. 7:581.

doi: 10.3389/fpsyg.2016.00581

Coworking Spaces: A Source ofSocial Support for IndependentProfessionalsCornelia Gerdenitsch1*, Tabea E. Scheel2, Julia Andorfer1 and Christian Korunka1

1 Department of Applied Psychology: Work, Education, Economy, Faculty of Psychology, University of Vienna, Vienna,Austria, 2 Social and Organizational Psychology, Department of Psychology, Faculty of Life Science, Humboldt University ofBerlin, Berlin, Germany

Coworking spaces are shared office environments for independent professionals. Suchspaces have been increasing rapidly throughout the world, and provide, in addition tobasic business infrastructure, the opportunity for social interaction. This article exploressocial interaction in coworking spaces and reports the results of two studies. Study1 (N = 69 coworkers) finds that social interaction in coworking spaces can take theform of social support. Study 2 further investigates social support among coworkers(N = 154 coworkers) and contrasts these results with those of social support amongcolleagues in traditional work organizations (N = 609). A moderated mediation modelusing time pressure and self-efficacy, based on the conservation of resources theory,is tested. Social support from both sources was positively related to performancesatisfaction. Self-efficacy mediated this relationship in the employee sample, while in thecoworking sample, self-efficacy only mediated the relationship between social supportand performance satisfaction if time pressure was high. Thus, a mobilization of socialsupport seems necessary in coworking spaces. We conclude that coworking spaces,as modern social work environments, should align flexible work infrastructure withwell-constructed opportunities for social support.

Keywords: coworking space, new ways of working, social support, entrepreneurship, ressources, socialinteraction

INTRODUCTION

Coworking spaces are shared office environments for independent professionals (Pohler, 2012;Spinuzzi, 2012) and have been increasing rapidly. In 2015, 7,800 spaces existed worldwide witha growth rate of 83% from 2012 to 2013 (Foertsch, 2013) and of 36% from 2014 to 2015 (Foertsch,2015). Various professionals, predominantly independent ones such as freelancers or remoteworkers, use these spaces as their places of business (Pohler, 2012). Most of these independentprofessionals worked from home prior to renting a place of work in a coworking space, wherethey may have suffered from feeling of isolation, among other problems (Spinuzzi, 2012). Thus,coworking spaces represent one possible buffer against isolation by providing, in addition tobusiness infrastructure, the opportunity for social interaction.

This paper reports the results of two studies. Within the first, we explore social interactions ina coworking space (Study 1) and question whether they can take the form of social support. Social

Frontiers in Psychology | www.frontiersin.org 1 April 2016 | Volume 7 | Article 581

fpsyg-07-00581 April 22, 2016 Time: 16:29 # 2

Gerdenitsch et al. Social Support in Coworking Spaces

support describes an exchange of resources between at leasttwo persons with the intention to help (House et al., 1988).Aspects of social support are direct support, affective support,or confirmation (Kahn and Antonucci, 1980; House et al.,1988). Besides examining whether social support is a reality incoworking spaces, we investigate the effect of social supportin a second study (Study 2). We focus on work-related socialsupport (Frese, 1989), which has been widely studied in thecontext of traditional workplace settings. Within the secondstudy, we particularly contrast the effect of social support fromcoworkers in these new spaces with that from colleagues intraditional workplace settings. We assume that coworkers in acoworking space may represent a new source of social supportfor independent professionals, operating along different linesthan colleagues and supervisors in a traditional setting. Wepropose a moderated mediation model of social support, whichwe expect to be valid for both the coworking and employeesamples. In particular, we expect a positive relationship of socialsupport to satisfaction with performance, mediated by self-efficacy (see also Rhoades and Eisenberger, 2002; Osca et al.,2005; Bakker and Demerouti, 2007). Further, we assume thatthis mediation is moderated by time pressure. We expect thatsupport will be activated when there is high time pressure, whichstrengthens mediation in both the traditional and coworkingsettings.

Although there has been an increase in the number ofcoworking spaces, scientific research has still not paidadequate attention to this emerging office environment.Qualitative research articles (Pohler, 2012; Spinuzzi, 2012;Garrett et al., 2014), case studies (Fuzi, 2015; Orel and Rus,2015), unpublished work (Foertsch, 2013), and blogs havedescribed the characteristics of coworking spaces, the peoplewho work there, reasons for working there, and how andwhen a sense of a community emerges in these spaces.However, to the best of our knowledge, there is no publishedempirical quantitative research article describing coworkingspaces and no study that explores social interactions and theeffects of being supported by other coworkers in coworkingspaces. This article contributes to research in this area in twoways.

First, within Study 1, we explore social interaction in acoworking space, considering the question of whether socialinteraction takes the form of social support. We thus extendexisting research on physical office environments (Elsbach andPratt, 2007; Ashkanasy et al., 2014) by investigating coworkingspaces as emerging new office environments. Second, existingresearch on social support in the work context has mainlyfocused on social support from colleagues and supervisors(Baruch-Feldman et al., 2002). We now argue that for thisspecific sample of independent professionals, alternative sourcesof social support may become more relevant and may bemore appropriate given that colleagues and supervisors areeither not consistently available or non-existent. Thus, ourstudy extends the research on social support by investigatingits presence in a new population: coworkers in a coworkingspace.

STUDY 1: SOCIAL INTERACTIONS INCOWORKING SPACES

Professionals in coworking spaces have flexibility regarding theirwork location and are thus described as the prototypes of the“boundaryless workforce” (Pohler, 2012). Such location flexibilitycomes with some obstacles. One is professional isolation (Vegaand Brennan, 2000; Bailey and Kurland, 2002), which “occurswhen the desires for support, understanding, and other social andemotional aspects of interaction are not met” (Taha and Caldwell,1993, p.277). In line with a recent article (Spinuzzi, 2012),we argue that coworking spaces provide a potential solutionto professional isolation, since they aim to establish a socialatmosphere in which social interaction and collaboration arepossible (Pohler, 2012; Moriset, 2014).

Social interactions in coworking spaces may come in variousforms. On the one hand, people may simply work alongsideeach other or engage in rather casual conversation. On theother hand, coworkers may engage in networking, seek andobtain feedback, share ideas, or collaborate (Spinuzzi, 2012).However, very little is known about what social interaction incoworking spaces looks like. Rather, it is still unclear whethersocial interaction in a coworking space takes the form of socialsupport, as it often does between colleagues in traditional worksettings.

Basically, social interaction is a process by which peopleact and react to those around them (Giddens, 2009). Socialinteractions take various forms that can be positive and/ornegative. The simple presence of others or the fact of workingwith others rather than alone can for instance be positive,although it does not yet represent socially supportive interactions.Social interactions that are beneficial to one or both partiesqualify as social support (Shinn et al., 1984). In particular, socialsupport describes an exchange of resources between at least twopersons, whereas the sender who provides support aims to helpthe person who receives the support (House et al., 1988). Threeaspects of social support are identified (Kahn and Antonucci,1980; House et al., 1988): direct support (instrumental support,exchange of information), affective support (admiration, liking),and confirmation about actions and statements.

Coworkers are independent workers and are thus not asclosely aligned as regular colleagues because they are usuallynot working toward the same goal. There is also no taskinterdependence between them. They may even be in direct opencompetition with one another if they work in a space that servespeople with a specific specialization or field of work. However,there are theoretical arguments that make us assume that socialsupport is a significant reality in coworking spaces. In referenceto the social identity approach, individuals classify themselvesas members of social categories to locate themselves within agiven environment (Tajfel and Turner, 1986; Asforth and Mael,1989). Coworkers may define themselves as part of the globalcoworking community group and/or as part of their specificworkspace group. Members of a group typically act in a way thatsupports the other members of this group (Asforth and Mael,1989).

Frontiers in Psychology | www.frontiersin.org 2 April 2016 | Volume 7 | Article 581

fpsyg-07-00581 April 22, 2016 Time: 16:29 # 3

Gerdenitsch et al. Social Support in Coworking Spaces

There are some indications that coworkers indeed identifyat least to some extent with the global coworking communityand/or the specific coworking space where they are working.People working in coworking spaces are part of the globalcoworking community, which is interconnected via diversemedia. For instance, a Coworking Wiki lists coworkingspaces worldwide and a definition about coworking spaces.This definition highlights that coworking spaces are morethan mere shared offices; coworkers share the same corevalues: collaboration, openness, community, accessibility, andsustainability1. Furthermore, conferences (global coworkingunconference conference; coworking Europe conference) bringtogether people of the community. On the level of the coworkingspace, “hosts” or “community managers” organize events tostrengthen exchange and community thinking. Garrett et al.(2014) also describe how a sense of community emerges incoworking spaces.

To sum up, the coworking movement includes some sort ofgroup thinking. Being part of the same social group promotessupportive behavior and makes it easier to ask coworkers to listento job-related as well as personal problems. Which may fostersupportive behavior. The aim of Study 1 is to explore the natureof social interactions that take place in coworking spaces. Weexpect that such social interactions can take the form of socialsupport.

Materials and MethodsSample and ProcedureThe sample consisted of 69 coworkers (average age = 32.02,SD= 5.99) working in eight different coworking spaces in Austria(45 male). Most (80%) held a university degree. The samplewas international, with most participants being Austrian (51%),followed by Germans (16%), and 18 other nationalities (30%;3% missing). We recruited the sample by personally promotingour study in coworking spaces in Austria and through socialmedia. The survey was available both online and in paper–pencilformat.

MeasuresThe survey consisted of sociodemographic questions (gender,age, tenure, nationality, education, and employment status)and questions pertaining to social interaction with othercoworkers. Specifically, we asked participants to describesituations in which they interacted with other coworkers.This approach of a retrospective self-report about specificsituations is similar to the critical incidence technique (Flanagan,1954; Butterfield et al., 2005) and enables us to gather richinformation about social interaction from the participants’perspectives. To ensure responses that describe a widespreadvariety of situations, we instructed participants to think ofthree situations, a casual/short interaction, medium-lengthinteraction, and a longer interaction. The question was wordedas follows: “Please think of 3 situations in your coworkingspace when you interacted with coworkers, one situation with ashort/casual social interaction, one with a medium length, and

1wiki.coworking.org

one situation with a long/intensive social interaction. Please brieflydescribe the situations in the following paragraphs (∼5 sentenceseach).”

AnalysesWe applied a summarizing qualitative content analysis (Mayring,2008) to cluster the situational descriptions. The first stepentailed reformulating original statements into a content-relatedlinguistic form (paraphrase). Three raters discussed the originalstatements and came up with a set of categories that describethe essence of the original statements. Subsequently, anotherindependent rater was asked to categorize the statementsdeductively.

ResultsWe collected 178 descriptions of social interactions in coworkingspaces (65 short, 58 medium, 55 long). Content analysis resultedin four categories representing the descriptions. Two representthe two aspects of direct social support that are instrumental,support and exchange of information. The others describeinformal social interaction and collaboration within projects.Cohen’s κ was run to determine agreement between the raters.There was a good degree of agreement (κ = 0.744, p < 0.0005)according to the threshold of 0.7 suggested by Landis and Koch(1977). More precisely, raters agreed with regard to the category“informal social interactions” in 74 (out of 85), “exchange ofinformation” in 33 (out of 41), “instrumental support” in 24(out of 30), and “collaboration” in 16 (out of 22) situations. Thecategories are described below.

Informal social interactions (85 statements). On a basic level,coworkers reported encounters when they greet other coworkersand have short conversations over coffee or cigarettes andover lunch. Two coworkers described the following situations.“After coming to the office in the morning, I usually get somecoffee. Other people are frequently passing by on their way in,so we typically have a short chit-chat” (#11, 31 years, male);“Was grabbing a coffee, met a coworker. Just briefly introducedourselves. Waited for his coffee to finish (coffee machine− > cup) as well. Said goodbye” (#53, 22 years, male). Thesesituations can also get deeper when coworkers have personalconversations about life while for instance going for drinks afterwork or doing sports together. One participant described thefollowing situation. “We were working late, and I didn’t care to gohome. I asked a colleague to go for a beer. We ended up talking forseveral hours and covered personal, work, and emotional topics”(#20, 25 years, male).

Exchange of information (41 statements). In contrast toinformal social interactions, this category includes socialinteractions that are explicitly work-related. Coworkers describework-related conversations with other coworkers, but also theirengagement in official networking activities in their coworkingspace. During lunchtime and coffee breaks, coworkers get toknow the projects other coworkers are working on. “When peoplehave lunch, we often sit and eat together while talking aboutprojects, tech-related things (with other engineers) or whatevercomes to mind. Most people know each other well enough thathaving lunch together is not awkward” (#11, 31 years, male).

Frontiers in Psychology | www.frontiersin.org 3 April 2016 | Volume 7 | Article 581

fpsyg-07-00581 April 22, 2016 Time: 16:29 # 4

Gerdenitsch et al. Social Support in Coworking Spaces

Besides updating each other on current projects and networking,coworkers also reported discussing potential collaborations orplanning common activities (workshops). They also reportedattending events or workshops organized in their coworkingspace: “I went to the AngularJS meetup. Listened to a couple ofinteresting lectures, met some interesting people. It was a goodopportunity to meet people with similar interests and learn a fewthings” (#7, 30 years, male).

Instrumental support (30 statements). Coworkers reportedasking for or providing help in terms of feedback, brainstorming,and coaching. In contrast to information exchange, thesestatements are about situations in which workers report helpingeach other in a concrete task. With regard to feedback, coworkersreported asking for feedback or providing support for problems.Examples of such feedback situations are the following: “Acoworker asked me for my opinion on some websites he wasdesigning” (#36, 35 years, male); “I asked a coworker about atechnical problem. He/she took 10 min of time to listen andpropose a solution” (#49, 53 years, male); “I was having a lunchsession with a coworker where I gave feedback on project idea”(#17, 30 years, male). Furthermore, coworkers asked others toengage in short brainstorming sessions, which can also takethe form of coaching. Three coworkers described the followingsituations: “I had a 1-h chat over coffee with a member of thecommunity about his next steps, as he’s at the moment standingon the crossroad of opportunities (changing career, discoveringwhere he wants to go); I also shared an idea with him aboutstarting” (#62, 25 years, male); “Business modeling support fora coworker’s start-up company, strategy to apply for publicgrant, strategy for talking to external investors/business angels,contact with and meetings with business angels” (#25, 43 years,male); “Meeting with a coworker every 14 days for mutualcoaching and exchange, for the past 2 years” (#51, 45 years,female).

Collaboration (22 statements). Besides providing feedback,brainstorming, and coaching, coworkers also engage incollaborations with one another, both paid and unpaid. Theyreported working together on an idea or ask others to take oversome tasks. One described recruiting someone in the space to dosome paid work: “A professional article was needed for a hugeonline magazine and a PR/marketing expert coworker helped usto write it” (#63, 33 years, female).

Brief Discussion of Study 1Study 1 improves the understanding about social interactionsthat occur in a coworking space. Coworkers describe mostof the situations as informal social interactions. We suggestthat those social interactions enable coworkers to create asocial network. Interestingly, we also found that collaborationbetween coworkers is possible. In these situations, peopleworking in coworking spaces managed to create synergies orcommon benefits between their own businesses in contrastto competing with each other. Finally, our findings showedthat situations of supportive behavior indeed take place incoworking spaces. The categories “instrumental support” and“exchange of information” represent aspects of direct socialsupport (Kahn and Antonucci, 1980; House et al., 1988).

Within Study 2, we aim to shed light on this form of socialinteractions. Work-related social support has been researchedas one relevant job resource in traditional workplace settings.The question remains if effects of social support from coworkersare similar to effects from colleagues in a traditional officesetting.

Limitations to the study involve possible self-selection ofparticipants due to the research approach. Coworkers who aremore integrated in their coworking space may be more willingto participate in such a study and more likely provide socialsupport. However, we do not conclude that every coworkerengages in social interaction, but demonstrate that social supportdoes indeed occur in coworking spaces. Thus, a coworking spacerepresents a social environment that provides social support,representing one important job resource.

STUDY 2: SOCIAL SUPPORT INCOWORKING SPACES

In the first study, we described that social interactions incoworking spaces take the form of direct social support, butthe effects of this social support are still unclear. Whileseveral studies have described beneficial effects of support fromcolleagues and supervisors on work stress (mitigated stressors,reduced strain, buffer between stressor and strain; meta-analysisby Viswesvaran et al., 1999) and job satisfaction (supervisorsupport leads to various beneficial outcomes mediated byperceived organizational support; meta-analysis by Rhoades andEisenberger, 2002) in traditional office settings, no study hasinvestigated the effects of social support from coworkers in acoworking space.

The aim of Study 2 was to contrast the effects of social supportin a traditional office and a coworking space setting. We expect tosee similar beneficial effects of social support from coworkers aswe do from colleagues in a traditional work setting. We collectedtwo samples, one in traditional office settings (colleagues assource of social support) and the other in coworking spaces(coworkers as source of social support) to contrast the effectsof social support from coworkers and from colleagues. Basedon the conservation of resources theory (COR theory; Hobfoll,1989, 2002), we derived a moderated mediation model that wetested for both sources of social support. The model is depicted inFigure 1. The specific hypotheses are described in the followingsections.

A considerable amount of literature has been publishedon the main and moderating effects of social support in theworking context (Cohen and Wills, 1985; Viswesvaran et al.,1999; Rhoades and Eisenberger, 2002). Several studies foundpositive effects of social support on performance (Searle et al.,2001; Baruch-Feldman et al., 2002; Osca et al., 2005) andgoal achievement (Bakker and Demerouti, 2007). In particular,social support can help employees feel good about themselves,leading to a positive evaluation of their performance (Searleet al., 2001). In line with this research, we assume thatexperiencing social support from colleagues positively affectsperformance.

Frontiers in Psychology | www.frontiersin.org 4 April 2016 | Volume 7 | Article 581

fpsyg-07-00581 April 22, 2016 Time: 16:29 # 5

Gerdenitsch et al. Social Support in Coworking Spaces

FIGURE 1 | Proposed research model.

For people working in coworking spaces, we expect tofind the same effect of social support. Spinuzzi (2012) arguesthat before visiting a coworking space, coworkers often sufferfrom professional isolation, which has a negative effect onperformance that can be buffered by, for example, moreface-to-face interaction (Golden et al., 2008). Therefore, weargue that coworkers will benefit from social support interms of performance. Since the population of coworkersis diverse with respect to their businesses (Pohler, 2012),a comparison of objective indicators of performances isdifficult, if not impossible. Therefore, we used self-ratedperformance satisfaction as a proxy measure for objective jobperformance.Hypothesis 1: We expect that social support from (a) colleaguesand (b) coworkers is positively related to self-rated performancesatisfaction.

Resource models propose that resources are linked to eachother, rather than existing in isolation (Hobfoll, 2002). Inthis vein, the beneficial effects of social support on work-related outcomes can be explained by a joint and reciprocalactivating effect of resources. COR theory (Hobfoll, 1989,2002) explains how resources relate to each other. This theorysuggests an enrichment of resources, a so-called resourcegain process (Hobfoll, 1989), whereby the existence of oneresource facilitates the development of other resources (Hobfoll,2002). As such, social support (one resource) can cleardoubts about one’s competence and thus increase self-efficacy(another resource). Self-efficacy describes the belief in one’sability to master challenges. It can be enhanced by socialpersuasion from significant others (Bandura, 1977). The roleof self-efficacy has been confirmed in several studies as amediator (Luthans et al., 2008; Mastenbroek et al., 2014)of the relationship between job resources and work-relatedoutcomes.

We expect that for employees as well as for coworkers, socialsupport activates self-efficacy in such a way that workers havemore faith in their own abilities to successfully master work-related challenges (increased self-efficacy). As a consequence,both employees and coworkers should be more satisfied withtheir performance.

Hypothesis 2: We expect that self-efficacy mediates therelationship between social support from (a) colleagues and(b) coworkers and performance satisfaction.

Conservation of resources theory suggests that resourcegain cycles are most likely to emerge during highly stressfulcircumstances (Hobfoll, 1989, 2002). The more stress,the more likely workers are to seek out or receive socialsupport (support-mobilization hypothesis, Eckenrode, 1983;Stephens and Long, 2000). Such a mobilization or activationof social support can again facilitate the development ofnew resources. However, a meta-analysis of employeedata found no evidence for the support-mobilizationhypothesis in the context of work stress (Viswesvaranet al., 1999). In the present study, we test this hypothesisin the context of performance. Drawing on COR theory,we expect that social support will be activated when timepressure is high, which will further facilitate development ofself-efficacy.

We argue that such an activation of social support is alsorelevant for professionals in coworking spaces. In contrast toemployees with colleagues and supervisors in traditional officecontexts, no predefined person is available for work-relatedquestions in a coworking space. Coworkers have to actively createand build a social network, and stressful situations with high timepressure can potentially stimulate support seeking. We furtherexpect that when social support is activated due to time pressure,the beneficial effects on satisfaction with performance via self-efficacy will also be strengthened (Bakker and Demerouti, 2007).Therefore, we formulated the following hypothesis.Hypothesis 3: We expect that time pressure moderates themediating effect of self-efficacy on the relationship between socialsupport from (a) colleagues and (b) coworkers and performancesatisfaction such that the mediating effect will be stronger whentime pressure is high.

Materials and MethodsSample and Procedure: ColleaguesThe employee sample consisted of 609 employees (330 female;average age = 29 years, SD = 3.90). In all, 35% obtained auniversity degree, and most of the coworkers worked full-time

Frontiers in Psychology | www.frontiersin.org 5 April 2016 | Volume 7 | Article 581

fpsyg-07-00581 April 22, 2016 Time: 16:29 # 6

Gerdenitsch et al. Social Support in Coworking Spaces

(90%). On average, they had worked 46.85 months (SD = 38.14)in their respective organizations. More than half of the sampleconsisted of Germans (53%), followed by Austrians (45%; 2%remaining). The sample of employees was gathered via an ISO-certified (ISO 26362) German online panel2, which ensureshigh quality data by minimizing participation frequency andconducting continuous controls.

Sample and Procedure: CoworkersThe coworking sample consisted of 154 coworkers acrossEurope (102 male and 52 female). Their average age was39 years (SD = 8.45). Most participants (79%) had a universitydegree, and 12% had a high school degree. Gender, age, andthe highest educational level of the sample were comparableto the non-scientific second annual coworking survey byDeskmag (N = 1532; Foertsch, 2012). In this study, thegender distribution was 66% male, 34% female, the averageage was 34, and 75% of the coworkers held a universitydegree. Participating coworkers had worked in a coworkingspace for an average of 18.1 months (SD = 22.2), whereasthey spent on average more than half of their working time(65%) in a coworking space. Altogether, the sample consisted ofcoworkers from 52 different coworking spaces from 17 countries,located in 37 cities. Participants included 24 nationalities, withthe majority being Austrian, Portuguese, and German (seeTable 1).

Participants were recruited using three strategies. First, wecontacted the managers of coworking spaces and asked themto distribute the online survey link to coworkers in theircoworking space and to send a reminder. To give potentialparticipants an understanding of the study and to motivatethem to take part, we prepared a self-made recruiting video.Of a list of 208 coworking spaces in 27 European countriesfrom the Coworking Wiki3, 50 spaces guaranteed their support.Second, we contacted journalists from relevant online magazines(Deskmag, Coworking News) and asked them to post thelink to the survey across a diverse array of social media andto write blog entries about the study. Finally, the study waspersonally promoted in coworking spaces in Vienna. Due tothe international sample, the online survey and all recruitingmaterials were in English.

MeasuresWe used the same measures for the sample of employeesand coworkers. To verify the adequacy of the survey for thesample of coworkers, we conducted a group discussion withthree members of a coworking space in Vienna (Austria) priorto the start of the study. The three members were a male(28 years) software developer, a female (30 years) owner of asocial media marketing agency, and a male (25 years) mobileapp developer. They filled out the first version of the survey,and we discussed the questionnaire’s items sequentially withregard to their comprehensibility and appropriateness for thetarget group of coworkers. The survey was modified to its final

2www.respondi.com3wiki.coworking.org

version taking their feedback into account. In particular, the twoitems for measuring performance satisfaction were created inconsultation with these members. The following variables (exceptthe questions about characteristics of the coworking sample) wereused to describe the working conditions of both employees andcoworkers.

Social supportSocial support was measured with the Work-Related SocialSupport Scale (SzSU, (Frese, 1989). Answers to the five questionswere rated from 1 (not at all) to 4 (completely). An example itemfrom this scale is: “How willing are these persons to listen to yourproblems with your job?” In answering the questions, employeeswere instructed to refer to colleagues and coworkers to othercoworkers.

Time pressureTime pressure was assessed using a four-item subscale from theInstrument for Stress-Related Job Analysis (ISTA, Semmer et al.,1998). The items were scored on a scale ranging from 1 (veryrarely/never) to 5 (very often). An example item is: “How often doyou have to work faster than normal in order to complete yourwork?”

Self-efficacySelf-efficacy was measured using four items from the generalizedself-efficacy scale (Schwarzer and Jerusalem, 1995). The fourstatements were rated from 1 (not at all true) to 4 (exactly true).An example item is: “I am confident that I can deal efficiently withunexpected events.”

Performance satisfactionPerformance satisfaction was captured with two self-developeditems measuring participants’ satisfaction with the quality of theirwork. The first item reads: “How satisfied are you with the qualityof your work?” The second item measures their satisfaction withachieving objectives: “How satisfied are you with the achievementof the goals you have set for your work?” The correlation betweenthe two items was r= 0.66 (p < 0.001) for employees and r= 0.67(p < 0.01) for coworkers. Answers were scored on a scale rangingfrom not at all (1) to extremely satisfied (5, respectively 7 inStudy 2).

Characteristics of the coworking sampleSince research on people working in coworking spaces is rare,we formulated questions to provide detailed information on thecoworking sample we recruited. The questions aimed to collectinformation on the following topics: a description about theprojects coworkers were currently working on (open answerformat), the name of the coworking space where participantswere working, its location (city), whether this coworking spacehad a community aspect (yes vs. no), and how long participantshad been working in the coworking space (months). We alsoasked participants how much of their work time they spentin several places (the coworking space, my office, a friend’soffice, a home office, other places) and the amount of timethey preferred to work in a coworking space (full-time [e.g.,9 to 5, 10 to 6, etc.], for a few hours a day, on weekends,at night, sporadically, other). Finally, we asked them why they

Frontiers in Psychology | www.frontiersin.org 6 April 2016 | Volume 7 | Article 581

fpsyg-07-00581 April 22, 2016 Time: 16:29 # 7

Gerdenitsch et al. Social Support in Coworking Spaces

TABLE 1 | Sample characteristics of the coworking sample.

Employment statusa % Nationality Frequencies

Self-employed 62 Austrian 50

Freelancer 35 Portuguese 20

Full-time employees 16 German 18

Part-time employees 9 Italian 10

Student 8 French 8

Other occupational contracts 7 Czech 5

Frequency of using a coworking spacea % Polish 5

Full-time 74 American 4

A few hours a day 29 Slovenian 4

Sporadically 12 British 3

On weekends 11 Bulgarian 3

At night 10 Hungarian 3

Other time preferences 8 Spanish 3

Occupation Frequencies US 2

Software/web development, design 27 Dutch 2

Consultancy, management 16 Mexican 2

Writing, journalism, blogging, language services 10 Brazilian 2

Science, research/technology, education 7 Others 8

Online (social media) marketing/communication, PR 7 Missing 2

Working for a space 5 Reasonsa %

Design, creative projects 5 Social interaction 83

Arts, architecture 3 Productivity 68

Social entrepreneurship 4 Networking 67

Tourism, gastronomy 2 The provision of infrastructure 66

Others 9 Flexibility 63

Missing 4 Workplaces usedb Mean (SD)

Coworking space 64.47 (29.05)

Home office 25.27 (24.19)

Other third places (café, on the move,etc.)

12.96 (13.19)

One’s own office 8.87 (20.73)

A friend’s office 2.49 (7.21)

aMultiple options possible. bReported means and standard deviations are related to the percentage of the respective working time at the different places.

chose to work in a coworking space and offered multipleanswers (structure in one’s work day, collaboration, flexibleworking, networking, social interaction, productivity, provisionof infrastructure, locational advantages, and cost-efficiency) thathad to be rated on a Likert scale from 1 (not important) to 5 (veryimportant).

Control variablesAge, gender, and tenure were measured and included in analysesas control variables.

ResultsAnalyses of the questions describing characteristics of coworkersshowed that the main reason for working at the coworking

space was to engage in social interactions (83%). Descriptivestatistics on coworkers’ employment status, frequency ofcoworking space use, occupation, used workplaces, nationality,and reasons for working in a coworking space are presentedin Table 1. Within Table 1 we also provide a comprehensivelist about projects coworkers are actually working at. Based onthe project descriptions one rater determined that coworkersworked primarily in the areas of “software/web developmentand design” (27%), “consultancy and management” (16%),or “writing, journalism, blogging, and language services”(10%). These categories were validated by a second raterwith no information about the study. The second raterassigned the project descriptions to one of these categories(κ= 0.72).

Frontiers in Psychology | www.frontiersin.org 7 April 2016 | Volume 7 | Article 581

fpsyg-07-00581 April 22, 2016 Time: 16:29 # 8

Gerdenitsch et al. Social Support in Coworking Spaces

TABLE 2 | Means, standard deviations, reliabilities (cronbach’s alpha on the diagonal), and correlations between the study variables.

M (SD) 1 2 3 4 5 6

Employee sample

(1) Age 27.66 (3.89) –

(2) Gender 1.46 (0.50) 0.11∗∗ –

(3) Tenure 46.85 (38.14) 0.37∗∗∗ 0.09∗ –

(4) Social support colleagues 3.06 (0.61) −0.03 −0.07 −0.08 (0.74)

(5) Self-efficacy 3.18 (0.52) 0.07 0.06 0.10∗ 0.25∗∗∗ (0.82)

(6) Time pressure 3.13 (0.90) 0.04 0.04 0.08 −0.06 0.08∗ (0.86)

(7) Performance satisfaction 4.92 (1.26) 0.05 0.07 0.07 0.24∗∗∗ −0.21∗∗∗ 0.30∗∗∗

Coworking sample

(1) Age 34.86 (8.45) –

(2) Gender 1.66 (0.48) −0.05 –

(3) Tenure 18.10 (22.31) 0.18∗ 0.11 –

(4) Social support coworkers 2.74 (0.58) −0.21∗ 0.06 0.09 (0.80)

(5) Self-efficacy 3.43 (0.43) 0.03 −0.02 0.06 0.10 (0.76)

(6) Time pressure 2.84 (0.75) 0.11 −0.07 −0.07 −0.01 0.19∗ (0.81)

(7) Performance satisfaction 3.93 (0.60) 0.10 −0.08 0.07 0.21∗ −0.13 0.17∗

Coworking sample N = 154. Employee sample N = 609. Gender: 1 = female, 2 = male. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.

Descriptive statistics, reliabilities (Cronbach’s alphas), andcorrelations among the study variables for both the employee andcoworking samples appear in Table 2. In all hypothesis testinganalyses, we controlled for age, gender, and tenure, which werenot significant in any of the regression models.

To examine the main effect of social support on performancesatisfaction (Hypothesis 1), we conducted hierarchical regressionanalyses. To test mediation (Hypotheses 2a and 2b) andmoderated mediation (Hypotheses 3a and 3b), we followthe procedure outlined by Preacher and Hayes (2004). Weuse an SPSS macro (process) to estimate both mediationand moderated mediation. This stepwise approach estimatesindirect effects with both the Sobel test and bootstrapping.Following Becker’s (2005) suggestions, we excluded the controlvariables from further analyses when they were not significantlycorrelated with the dependent variable in the regressionmodel. We plotted simple slopes to interpret interaction effectsat one standard deviation above and below the mean ofthe moderator (Aiken and West, 1991). In the coworkingsample, we replaced 0.7% (16 out of 2,310) missing values.As recommended by several authors (e.g., Acock, 2005), weensured that the data were missing completely at random withLittle’s Missing Completely at Random (MCAR) test [χ2(361,N = 154) = 366.30, p = 0.41], and we replaced missingdata using the expectation-maximization (EM) algorithm inSPSS.

The testing of the postulated hypotheses showed the following:first, we tested the direct effect of social support on performancesatisfaction and found a significant positive relationship betweenthese two variables in the employee sample (β = 0.24, p < 0.001;1R2= 0.06, 1F = 37.34, p < 0.001) as well as in the coworker

sample (β= 0.21, p= 0.010; 1R2= 0.04, 1F = 6.80, p= 0.010).

Results indicate that workers experiencing social support fromtheir colleagues or coworkers are more satisfied with their

performance in terms of work quality and achieving objectives.Thus, Hypotheses 1a and 1b are supported.

Second, we tested whether the relationship between socialsupport and performance satisfaction was mediated by self-efficacy (see Table 3). For the employee sample, the effects ofsocial support on self-efficacy (path a; β = 0.25, p < 0.001)and of self-efficacy on satisfaction with performance (path b;β = 0.26, p < 0.001) were significant. Also, the total effect fromsocial support to performance satisfaction was significant (path c:β= 0.24, p < 0.001) and remained significant although weakenedwhen the mediator was included (path c’: β = 0.18, p < 0.001).These findings were supported by a test of the indirect effect viabootstrapping (95% CI [0.07, 0.21]).

For the coworking sample, the effect of social support onself-efficacy (path a; p = 0.197) and the effect of self-efficacyon satisfaction with performance (path b; β = 0.15, p = 0.062)were not significant. The total effect of social support onperformance satisfaction (path c) was positive and significant(β= 0.21, p= 0.010) and remained significant when self-efficacywas included as a mediator (path c’: β = 0.19, p = 0.017).These findings were supported by a test of the indirect effectvia bootstrapping (95% CI [−0.00, 0.08]). Thus, these resultssuggest that self-efficacy mediates the effect of social support onperformance satisfaction for employees, but not for coworkersoverall, supporting Hypothesis 2a (employee sample), but notHypothesis 2b (coworking sample).

Finally, we tested whether the mediating effect of self-efficacywas moderated by time pressure according to the analyticalapproach suggested by Preacher et al. (2007). We do not findsuch a moderated mediating effect for the employee sample(90% CI [−0.07, 0.05]). However, in the coworking sample,the conditional mediation of self-efficacy was stronger andsignificant at higher (90% CI [0.01, 0.13]) and medium (90%CI [0.00, 0.07]) levels of time pressure and was weaker and

Frontiers in Psychology | www.frontiersin.org 8 April 2016 | Volume 7 | Article 581

fpsyg-07-00581 April 22, 2016 Time: 16:29 # 9

Gerdenitsch et al. Social Support in Coworking Spaces

TABLE 3 | Results for testing mediation of self-efficacy.

Pathway Coworking Sample Employee Sample

β B SE t P β B SE t p

Path a (Social Support −> Self-efficacy) 0.25 0.21 0.03 6.40 <0.001 0.10 0.08 0.06 1.30 0.20

Path b (Self-efficacy –> PerformanceSatisfaction)

0.26 0.63 0.10 6.54 <0.001 0.15 0.20 0.11 1.81 0.07

Total effect (Path c, SocialSupport – > Performance Satisfaction)

0.24 0.50 0.08 6.11 <0.001 0.21 0.21 0.08 2.61 0.01

Direct effect (Path c’, Social Support onPerformance Satisfaction includingSelf-efficacy)

0.18 0.36 0.08 4.47 <0.001 0.19 0.19 0.08 2.42 0.01

Effect Boot SE LLCI ULCI Effect Boot SE LLCI ULCI

Indirect effect (Paths a × b) of SocialSupport on Performance Satisfactionvia Self-efficacy

0.13 0.03 0.07 0.21 0.02 0.02 −0.00 0.08

Coworking Sample N = 154. Employee Sample N = 609. LL, lower limit; UL, upper limit; CI, confidence interval; number of bootstraps: 100.

became non-significant at lower levels of time pressure (90%CI = [−0.06, 0.00]). To further clarify this moderating effect,we examined separate simple slopes (Aiken and West, 1991).For the low (b = 0.004, SE = 0.05, p = 0.942) and medium(b= 0.081, SE= 0.05, p= 0.142) time pressure groups, the simpleslopes were not significant, whereas the simple slopes for the high(b= 0.158, SE= 0.06, p= 0.005) time pressure group was positiveand significant (see Figure 2). Thus, Hypothesis 3 was supportedin the coworking sample, but not in the employee sample.

Brief Discussion of Study 2Findings of Study 2 showed both similar and different effectsof social support from colleagues in a traditional office settingand coworkers in a coworking space. First, consistent withprevious literature (Searle et al., 2001; Baruch-Feldman et al.,2002; Osca et al., 2005), we found a positive main effect ofsocial support from colleagues and coworkers on satisfactionwith performance. Regarding the hypotheses derived from CORtheory, we found different effects for employees and coworkers.The proposition of a mediation of self-efficacy (resource gainprocess) found support only in the employee sample. In thecoworking sample, the mediation was only significant in thehigh time pressure group, providing evidence for the support-mobilization hypothesis (Eckenrode, 1983). In line with previousempirical evidence (Viswesvaran et al., 1999), we found noconfirmation for the support-mobilization hypothesis in theemployee sample. We conclude that the mobilization of resourcesseems to be more necessary in coworking spaces than intraditional working contexts.

The main strength of the second study is its inclusion ofcoworkers working in various coworking spaces across Europe.Moreover, it is the first quantitative multinational study toinvestigate the effects of social support in coworking spaces.Results are more generalizable than those of previous studiesbecause our study was not limited to one country (see Spinuzzi,

2012) or a small selection of coworking spaces (see Pohler,2012). However, as the selection of participants is not random,generalizability on the population of coworkers in coworkingspaces is limited.

The main limitation of Study 2 concerns the comparabilityof the two samples. However, the two samples mirror theirrespective populations well. Another limitation concernsthe cross-sectional design, which does not allow for causalinferences. As such, it is impossible to conclude with certaintythat social support causes performance satisfaction and not viceversa. However, COR theory suggests a direct effect from theresource social support to consequences. Furthermore, as thedata used are based on self-reports, common method bias ispossible (Podsakoff et al., 2012). Following recommendationsof Podsakoff et al. (2003), we used different response formatsand valid scales to reduce possible common method variance.Nevertheless, we did not temporally separate the predictorand criterion. However, since research on coworking spacesis rare, our studies provide valuable information on whichto build in the future. Our findings highlight relationshipsbetween social support, self-efficacy, and performancesatisfaction and emphasize the role of coworking spaces as socialecosystems.

DISCUSSION ON FINDINGS OF STUDIES1 AND 2

Coworking spaces are office environments for independentprofessionals and are rapidly spreading worldwide. One mainreason professionals opt to work in such spaces is the opportunityfor social interaction, which diminishes the isolation independentprofessionals often struggle with (Spinuzzi, 2012). Findingsof Study 1 showed that these social interactions come invarious forms such as informal social interactions, direct socialsupport (instrumental support, exchange of information), and

Frontiers in Psychology | www.frontiersin.org 9 April 2016 | Volume 7 | Article 581

fpsyg-07-00581 April 22, 2016 Time: 16:29 # 10

Gerdenitsch et al. Social Support in Coworking Spaces

FIGURE 2 | The moderating effect of time pressure on the relationbetween social support from coworkers and self-efficacy.

collaboration. Thus, our findings indicate that coworking spacesare social environments that can provide possibilities for socialsupport with coworkers as a new source of social support.Study 2 showed the effects of this social support and contrastedthem with the effects in a traditional work setting. Interestingly,we found a moderated mediation in the coworking sample,but not in the sample of traditional working employees. Itseems that a mobilization of support (when time pressure ishigh) is relevant in the coworking sample. We also foundin Study 1 that coworkers engage in various informal socialinteractions, which can function as a precondition of socialsupport. Engaging in informal social interactions and mobilizingsupport may cost energy (Norris and Kaniasty, 1996) and shouldtherefore be facilitated by the management of the coworkingspace.

Implications for PracticeThe findings of the present study underpin the importanceof social support in coworking spaces and should encouragecoworking spaces to provide the types of contingenciesthat facilitate social support. Coworking spaces can, forexample, display information about other currently presentcoworkers. Such displays can provide an icebreaker forconversations (Bilandzic et al., 2013). Starting conversationsmay even be easier when such displays include informationabout coworkers’ backgrounds, skills, or availability. By suchmeans, coworking spaces can establish interaction as a socialnorm. Furthermore, we recommend specialization when spaceswant to increase social support. Like-minded people orpeople with similar occupations are more likely to be insimilar situations and can consequently better support eachother.

At this point, we would like to point out that besidesfacilitating social support, coworking spaces should also fulfillother needs of coworkers for concentrated work or havingdedicated space to perform individual work activities. Quietrooms in which people can concentrate are as important asgroup workspaces (Seddigh et al., 2014), especially when socialinteractions may distract other coworkers.

Limitations and Suggestions for FutureResearchWithin the present article, we were able to describe forms ofsocial interactions that happen in coworking spaces. We theninvestigated effects of one of these forms, social support, butnot of the others. Therefore, we encourage further studies toinvestigate the antecedents and consequences of the two otherforms of social interactions in coworking spaces we found inStudy 1. These studies may also consider moderating variablessuch as spatial needs or personal preferences of coworkers. Forinstance, informal social interactions may diminish feelings ofisolation but can also distract coworkers when they need toconcentrate on working tasks. Collaboration, on the other hand,may improve productivity of the owned business. Personalityvariables may also have an impact here. For instance, for peoplehigh in extraversion, a first contact may be easier. We alsoencourage further studies to consider not only social interactionbut also aspects of the social atmosphere. For instance, the merepresence of other people can also have an effect on coworkers.Due to social facilitation, simple tasks are performed better in thepresence of others (Zajonc, 1965).

The two studies presented within this article represent animportant step in describing the social aspect in coworkingspaces in general. However, it is still unclear why and underwhich circumstances coworkers interact with each other. Furtherstudies should deepen the understanding about interactions incoworking spaces. Further studies can, for instance, focus oncoworkers’ motives for and personal preferences in interactingwith others. Narrative qualitative approaches and analyticalapproaches that discover aspects below the direct conversationcan be useful here to shed light on why coworkers engagein social interactions. It would also be interesting to combinethis qualitative approach with network analysis investigating thedensity and strengths of ties between coworkers in a coworkingspace and to compare resulting networks with workers in atraditional workplace. Supervisors and colleagues as sources ofsocial support are predefined in traditional workplace settings. Inaddition, workers may have more superficial or other (private)relationships with other workers of the organization. Peopleworking in coworking spaces engage in various forms of socialinteractions without predefined others to ask for support.Furthermore, feedback and input from other coworkers can bemore easily ignored than those from traditional workers. A betterunderstanding of motives, preferences, and networks will enablethe implementation of targeted interventions to increase socialsupport in coworking spaces.

The cross-sectional approach of Study 2 is limited becauseit does not allow causal conclusions. Thus, there is a needfor longitudinal studies that examine predictors, activators, andconsequences in the context of receiving support in a coworkingspace. After receiving social support, various effects are possible.For instance, according to the social support deteriorationdeterrence model (Norris and Kaniasty, 1996), deteriorationoccurs after a mobilization of resources. Therefore, we can alsothink of negative effects such as emotional exhaustion in thelong term. There may also be differences between traditional

Frontiers in Psychology | www.frontiersin.org 10 April 2016 | Volume 7 | Article 581

fpsyg-07-00581 April 22, 2016 Time: 16:29 # 11

Gerdenitsch et al. Social Support in Coworking Spaces

office workers and coworkers. Coworkers may have to put moreeffort in activating support, while at the same time they canignore feedback more easily than workers can from colleaguesand supervisors. We further recommend for these studies to relyon data that are not solely self-reported to avoid common methodbias. For instance, social support can be assessed from differentsources, or objective or physiological data about emotionalexhaustion can be included.

CONCLUDING REMARKS

In the last couple of years, increasing numbers of independentprofessionals have opted to work in coworking spaces.This emerging office type appears to provide a resourcefulenvironment for this particular target group because it providesopportunities for social support in addition to flexible businessinfrastructure. To date, only a few scientific investigations ofcoworking spaces have been conducted. Our second studyis, to the best of our knowledge, the first to quantitativelyinvestigate social support in coworking spaces across Europe.The findings highlight the importance of coworkers as a sourceof social support among independent professionals and shouldencourage studies that further explore coworking spaces as

a social office environment likely to grow even more in thefuture.

AUTHOR CONTRIBUTIONS

CG, TES, and CK designed Study 1; CG, JA, and CK designedStudy 2; CG and JA conducted research; CG, JA, and TESanalyzed the data; CG wrote the article; TES edited the article;All authors read and edited a draft of the final article.

FUNDING

Data collection of the employee sample was conduced withfunding from the Austrian chamber of commerce. This articlewas supported by the Open Access Publishing Fund of theUniversity of Vienna.

ACKNOWLEDGMENTS

A special thanks for their support goes to all participatingCoworking Spaces (especially Sektor5 in Vienna), participatingcoworkers, Deskmag, and Coworking News.

REFERENCESAcock, A. C. (2005). Working with missing values. J. Marriage Fam. 67, 1012–1028.

doi: 10.1111/j.1741-3737.2005.00191.xAiken, L. S., and West, S. G. (1991). Multiple Regression: Testing and Interpreting

Interaction. Thousand Oaks, CA: Sage.Asforth, B. E., and Mael, F. (1989). Social identity theory and the organization.

Acad. Manag. Rev. 14, 20–39. doi: 10.5465/AMR.1989.4278999Ashkanasy, N. M., Ayoko, O. B., and Jehn, K. A. (2014). Understanding the physical

environment of work and employee behavior: an affective events perspective.J. Organ. Behav. 35, 1169–1184. doi: 10.1002/job.1973

Bailey, D. E., and Kurland, N. B. (2002). A review of telework research: findings,new directions, and lessons for the study of modern work. J. Organ. Behav. 23,383–400. doi: 10.1002/job.144

Bakker, A. B., and Demerouti, E. (2007). The job demands-resources model:State of the art. J. Manag. Psychol. 22, 309–328. doi: 10.1108/02683940710733115

Bandura, A. (1977). Self-efficacy: toward a unifying theory of behavioral change.Psychol. Rev. 84, 191–215. doi: 10.1037/0033-295X.84.2.191

Baruch-Feldman, C., Brondolo, E., Ben-Dayan, D., and Schwartz, J. (2002). Sourcesof social support and burnout, job satisfaction, and productivity. J. Occup.Health Psychol. 7, 84–93. doi: 10.1037//1076-8998.7.1.84

Becker, T. E. (2005). Potential problems in the statistical control of variables inorganizational research: a qualitative analysis with recommendations. Organ.Res. Methods 8, 274–289. doi: 10.1177/1094428105278021

Bilandzic, M., Schroeter, R., and Foth, M. (2013). “Gelatine: making coworkingplaces gel for better collaboration and social learning,” in Proceedings ofthe 25th Australian Computer-Human Interaction Conference: Augmentation,Application, Innovation, Collaboration (OzCHI), eds H. Shen, R. Smith,J. Paay, P. Calder, and T. Wyeld (New York, NY: ACM Digital Library),427–436.

Butterfield, L. D., Borgen, W. A., Amundson, N. E., and Maglio, A.-S. T. (2005).Fifty years of the critical incident technique: 1954-2004 and beyond. Qual. Res.5, 475–497. doi: 10.1177/1468794105056924

Cohen, S., and Wills, T. A. (1985). Stress social support and the bufferinghypothesis. Psychol. Bull. 98, 310–357. doi: 10.1037/0033-2909.98.2.310

Eckenrode, J. (1983). The mobilization of social supports: some individualconstraints. Am. J. Community Psychol. 11, 509–528. doi: 10.1007/BF00896802

Elsbach, K. D., and Pratt, M. G. (2007). The physical environment in organizations.Acad. Manag. Ann. 1, 181–224. doi: 10.1080/078559809

Flanagan, J. C. (1954). The critical incident technique. Psychol. Bull. 51, 327–358.doi: 10.1037/h0061470

Foertsch, C. (2012). 1st Results of the 3rd Global Coworking Survey. Deskmag,November 8, 2012. Available at: http://www.deskmag.com/en/1st-results-of-the-3rd-global-coworking-survey-2012 [accessed January 12, 2016]

Foertsch, C. (2013). 4.5 New Coworking Spaces Per Work Day. Deskmag, March 4,2013. Available at: http://www.deskmag.com/en/2500-coworking-spaces-4-5-per-day-741 [accessed January 12, 2016]

Foertsch, C. (2015). First Results of the New Global Coworking Survey. Deskmag,November 20, 2013. Available at: http://www.deskmag.com/en/first-results-of-the-new-global-coworking-survey-2015-16 [accessed January 12, 2016]

Frese, M. (1989). SzSU - skala zur sozialen unterstützung am arbeitsplatz [work-related social support scale]. Z. Arbeitswissenschaft 43, 112–121.

Fuzi, A. (2015). Co-working spaces for promoting entrepreneurship in sparseregions: the case of South Wales. Reg. Stud. Reg. Sci. 2, 462–469. doi:10.1080/21681376.2015.1072053

Garrett, L. E., Spreitzer, G. M., and Bacevice, P. (2014). Co-constructing a sense ofcommunity at work: the emergence of community in coworking spaces. Acad.Manag. Annu. Meet. Proc. 1, 14004–14004. doi: 10.5465/AMBPP.2014.139

Giddens, A. (2009). Sociology. Cambridge: Polity Press.Golden, T. D., Veiga, J. F., and Dino, R. N. (2008). The impact of

professional isolation on teleworker job performance and turnover intentions:does time spent teleworking, interacting face-to-face, or having access tocommunication-enhancing technology matter? J. Appl. Psychol. 93, 1412–1421.doi: 10.1037/a0012722

Hobfoll, S. E. (1989). Conservation of resources: a new attempt atconceptualizing stress. Am. Psychol. 44, 513–524. doi: 10.1037/0003-066X.44.3.513

Hobfoll, S. E. (2002). Social and psychological resources andadaptation. Rev. Gen. Psychol. 6, 307–324. doi: 10.1037//1089-2680.6.4.307

Frontiers in Psychology | www.frontiersin.org 11 April 2016 | Volume 7 | Article 581

fpsyg-07-00581 April 22, 2016 Time: 16:29 # 12

Gerdenitsch et al. Social Support in Coworking Spaces

House, J. S., Umberson, D., and Landis, K. R. (1988). Structure andprocess of social support. Annu. Rev. Sociol. 14, 293–318. doi:10.1146/annurev.so.14.080188.001453

Kahn, R. L., and Antonucci, T. (1980). “Convoys over the life course: attachment,roles and social support,” in Life Span Development and Behavior, eds P. B. Baltesand O. Brim (Boston, MA: Lextington Press), 254–283.

Landis, J. R., and Koch, G. G. (1977). The measurement of observer agreement forcategorical data. Biometrics 33, 159–174. doi: 10.2307/2529310

Luthans, F., Norman, S. M., Avolio, B. J., and Avey, J. B. (2008). The mediatingrole of psychological capital in the supportive organizational climate-employeeperformance relationship. J. Organ. Behav. 29, 219–238. doi: 10.1002/job.507

Mastenbroek, N. J. J. M., Jaarsma, A. D. C., Scherpbier, A. J. J. A., van Beukelen, P.,and Demerouti, E. (2014). The role of personal resources in explaining well-being and performance: a study among young veterinary professionals. Euro. J.Work Organ. Psychol. 23, 190–202. doi: 10.1080/1359432X.2012.728040

Mayring, P. (2008). Qualitative Inhaltsanalyse: Grundlagen und Techniken[Qualitative Content Analysis: Basics and Techniques]. Weinheim: DeutscherStudien Verlag.

Moriset, B. (2014). Building new places of the creative economy. The riseof coworking spaces. Paper Presented at the 2nd Geography of InnovationInternational Conference 2014. Utrecht.

Norris, F. H., and Kaniasty, K. (1996). Received and perceived social support intimes of stress: a rest of the social support deterioration deterrence model.J. Pers. Soc. Psychol. 71, 498–511. doi: 10.1037/0022-3514.71.3.498

Orel, M., and Rus, A. (2015). Coworking: a community of work. Teorija Praksa 52,1017–1038.

Osca, A., Urien, B., González-Camino, G., Dolores Martínez-Pérez, M., andMartínez-Pérez, N. (2005). Organisational support and group efficacy: alongitudinal study of main and buffer effects. J. Manag. Psychol. 20, 292–311.doi: 10.1108/02683940510589064

Podsakoff, P. M., MacKenzie, S. B., Lee, J.-Y., and Podsakoff, N. P. (2003). Commonmethod biases in behavioral research: a critical review of the literature andrecommended remedies. J. Appl. Psychol. 88, 879–903. doi: 10.1037/0021-9010.88.5.879

Podsakoff, P. M., MacKenzie, S. B., and Podsakoff, N. P. (2012). Sources of methodbias in social science research and recommendations on how to control it. Annu.Rev. Psychol. 63, 539–569. doi: 10.1146/annurev-psych-120710-100452

Pohler, N. (2012). Neue arbeitsräume für neue arbeitsformen: coworking spaces[New workspaces for new forms of work: coworking spaces]. Österr. Z. Soziol.37, 65–78. doi: 10.1007/s11614-012-0021-y

Preacher, K. J., and Hayes, A. F. (2004). SPSS and SAS procedures for estimatingindirect effects in simple mediation models. Behav. Res. Methods Instrum.Comput. 36, 717–731. doi: 10.3758/BF03206553

Preacher, K. J., Rucker, D. D., and Hayes, A. F. (2007). Addressing moderatedmediation hypotheses: theory, methods, and prescriptions. Multivar. Behav.Res. 42, 185–227. doi: 10.1080/00273170701341316

Rhoades, L., and Eisenberger, R. (2002). Perceived organizational support: a reviewof the literature. J. Appl. Psychol. 87, 698–714. doi: 10.1037//0021-9010.87.4.698

Schwarzer, R., and Jerusalem, M. (1995). “Generalized self-efficacy scale,” inMeasures in Health Psychology: A User’s Portfolio. Casual and Control Beliefs, edsJ. Weinman, S. Wright, and M. Johnston (London: NFER-NELSON), 35–37.

Searle, B., Bright, J. E. H., and Bochner, S. (2001). Helping people to sort it out:the role of social support in the job strain model. Work Stress 15, 328–346. doi:10.1080/02678370110086768

Seddigh, A., Berntson, E., Bodin Danielson, C., and Westerlund, H. (2014).Concentration requirements modify the effect of office type on indicatorsof health and performance. J. Environ. Psychol. 38, 167–174. doi:10.1016/j.jenvp.2014.01.009

Semmer, N. K., Zapf, D., and Dunckel, H. (1998). Instrument Zur StreßbezogenenTätigkeitsanalyse ISTA [Instrument for Stress-Related Job Analysis ISTA]. Zürich:Verlag der Fachvereine Hochschulverlag.

Shinn, M., Lehmann, S., and Wong, N. W. (1984). Social interaction and socialsupport. J. Soc. Issues 40, 55–76. doi: 10.1111/j.1540-4560.1984.tb01107.x

Spinuzzi, C. (2012). Working alone together: coworking as emergent collaborativeactivity. J. Bus. Tech. Commun. 26, 399–441. doi: 10.1177/1050651912444070

Stephens, C., and Long, N. (2000). Communication with police supervisors andpeers as a buffer of work-related traumatic stress. J. Organ. Behav. 21, 407–424.doi: 10.1002/(SICI)1099-1379(200006)21:4<407::AID-JOB17>3.0.CO;2-N

Taha, L. H., and Caldwell, B. S. (1993). Social isolation and integrationin electronic environments. Behav. Inform. Technol. 12, 276–283. doi:10.1080/01449299308924391

Tajfel, H., and Turner, J. C. (1986). “The social identity theory of intergroupbehaviour,” in Psychology of Intergroup Relations, eds S. Worchel and W. G.Austrin (Chicago, IL: Nelson-Hall), 7–24.

Vega, G., and Brennan, L. (2000). Isolation and technology: the human disconnect.J. Organ. Change Manag. 13, 468–481. doi: 10.1108/09534810010377435

Viswesvaran, C., Sanchez, J. I., and Fisher, J. (1999). The role of social support inthe process of workstress: a meta-analysis. J. Vocat. Behav. 54, 314–334. doi:10.1006/jvbe.1998.1661

Zajonc, R. B. (1965). Social facilitation. Science 149, 269–274. doi:10.1126/science.149.3681.269

Conflict of Interest Statement: The authors declare that the research wasconducted in the absence of any commercial or financial relationships that couldbe construed as a potential conflict of interest.

Copyright © 2016 Gerdenitsch, Scheel, Andorfer and Korunka. This is an open-accessarticle distributed under the terms of the Creative Commons Attribution License(CC BY). The use, distribution or reproduction in other forums is permitted, providedthe original author(s) or licensor are credited and that the original publication in thisjournal is cited, in accordance with accepted academic practice. No use, distributionor reproduction is permitted which does not comply with these terms.

Frontiers in Psychology | www.frontiersin.org 12 April 2016 | Volume 7 | Article 581